Genome-Wide Association Mapping for Agronomic and Seed Quality Traits of Field Pea (Pisum sativum L.)

Abstract

Genome-wide association study (GWAS) was conducted to identify loci associated with agronomic (days to flowering, days to maturity, plant height, seed yield and seed weight), seed morphology (shape and dimpling), and seed quality (protein, starch, and fiber concentrations) traits of field pea (Pisum sativum L.). A collection of 135 pea accessions from 23 different breeding programs in Africa (Ethiopia), Asia (India), Australia, Europe (Belarus, Czech Republic, Denmark, France, Lithuania, Netherlands, Russia, Sweden, Ukraine and United Kingdom), and North America (Canada and USA), was used for the GWAS. The accessions were genotyped using genotyping-by-sequencing (GBS). After filtering for a minimum read depth of five, and minor allele frequency of 0.05, 16,877 high quality SNPs were selected to determine marker-trait associations (MTA). The LD decay (LD_1/2max,90) across the chromosomes varied from 20 to 80 kb. Population structure analysis grouped the accessions into nine subpopulations. The accessions were evaluated in multi-year, multi-location trials in Olomouc (Czech Republic), Fargo, North Dakota (USA), and Rosthern and Sutherland, Saskatchewan (Canada) from 2013 to 2017. Each trait was phenotyped in at least five location-years. MTAs that were consistent across multiple trials were identified. Chr5LG3_566189651 and Chr5LG3_572899434 for plant height, Chr2LG1_409403647 for lodging resistance, Chr1LG6_57305683 and Chr1LG6_366513463 for grain yield, Chr1LG6_176606388, Chr2LG1_457185, Chr3LG5_234519042 and Chr7LG7_8229439 for seed starch concentration, and Chr3LG5_194530376 for seed protein concentration were identified from different locations and years. This research identified SNP markers associated with important traits in pea that have potential for marker-assisted selection towards rapid cultivar improvement.

Keywords: field pea; genetic diversity; genome-wide association study; genotyping-by-sequencing; single nucleotide polymorphisms.

Figure 1

Distribution of SNP markers selected for population structure and trait association analysis across the seven chromosomes of pea. The graph represents number of SNPs in each million bp of genetic distance of the seven pea chromosomes. The chromosome and linkage group assignment was in accordance to the pea genome assembled by Kreplak et al. (2019). The graphs are based on number of SNPs identified on chromosomes 1 to 7 (1685, 1768, 1786, 2356, 2917, 2349 and 2747, respectively).

Figure 2

Chromosome-wise linkage disequilibrium decay based on 135 pea accessions. The decline of LD- r² between SNPs pairs is presented as a function of physical distance in base pairs.

Figure 3

Population structure of 135 pea accessions based on K = 9. In the panel, each accession is indicated as a vertical bar partitioned into colored segments where the respective length of these segments represents the proportion of the individual's genome in a given group.

Figure 4

Genetic relatedness among the 135 pea accessions estimated by neighbor-joining method and represented as a polar tree diagram. The estimated genetic relatedness is based on 16,877 SNPs identified by genotyping-by-sequencing and filtered for minor allele frequency of 0.05.

Figure 5

Manhattan plots and the corresponding Q-Q plots representing the identification of SNP markers associated with plant height in multiple trials. (A) 2013 Fargo (B) 2013 Sutherland, (C) 2014 Sutherland, (D) 2015 Sutherland, (E) 2016 Rosthern, (F) 2016 Sutherland, (G) 2017 Rosthern, and (H) 2017 Sutherland. The Manhattan plots are based on association of 15608 chromosomal and1269 non-chromosomal SNPs with plant height of 135 pea accessions in the multi-year, multi-locational trials.

Figure 6

Manhattan plots and the corresponding Q-Q plots representing the identification of SNP markers associated with lodging resistance in multiple trials. (A) 2013 Fargo, (B) 2014 Fargo, (C) 2014 Sutherland, (D) 2015 Fargo, (E) 2015 Sutherland, (F) 2016 Rosthern, (G) 2016 Sutherland, and (H) 2017 Sutherland. The Manhattan plots are based on association of 15608 chromosomal and 1269 non-chromosomal SNPs with the lodging score measured on a 1-9 rating scale (1 = upright to 9 = completely lodged) in 135 pea accessions in the multi-year, multi-locational trials.